Defining Core Concepts and Measurements in Suicidology

Defining Core Concepts and

Measurements in SuicidologyBy: Lars Mehlum, University of Oslo & Jill

Harkavy-Friedman, AFSP

Lars MehlumProfessor of psychiatry and suicidologyDirector National Centre for Suicide Research and PreventionFaculty of MedicineUniversity of Oslo, OsloNorway

Immediate Past President IASR

Outline• Clarity of concepts and definitions – why is this important?

• Defining suicide, attempted suicide, self-harm and non-suicidal self-injury

• Some examples of instruments to measure these behaviours

• More core concepts in suicide research and how to measure

• Repeated measures, time windows

• Ecological momentary assessment

Many studies of suicidal behaviour fail to provide clear definitions of what they are studying

• Creates big problems for referees, readers and for those who want to conduct systematic literature reviews and meta-analyses

• This is a waste of resources and impedes scientific progress• Core concepts in any study should be clearly defined• If you are going to study suicidal behaviour, you should describe and

define these behaviours clearly and in behavioural terms• Far too many ways of defining suicidal behaviours already exist, so

unless you are making development of definitions the focus of your research, don’t add yet another one

Outline• Clarity of concepts and definitions – why is this important?

• Defining suicide, attempted suicide, self-harm and non-suicidal self-injury

• Some examples of instruments to measure these behaviours

• More core concepts in suicide research and how to measure

• Repeated measures, time windows

• Ecological momentary assessment

Self-Harm (SH)

Non-Suicidal Self-Injury NSSI

Suicidal Self-Harm

SuicideSuicideAttempt

Definitions• Self-harm

Any act of self-poisoning or self-injury carried out by an individual irrespective of motivation

• Suicide

Death caused by injuring oneself with the intent to die

• Suicide attempt

A potentially self-injurious act carried out with at least some wish to die, as a result of

act. There does not have to be any injury or harm, just the potential

• Non-suicidal self-injury

Intentional destruction of one’s own body tissue without suicidal intent and for purposes not

socially sanctioned

Outline• Clarity of concepts and definitions – why is this important?

• Defining suicide, attempted suicide, self-harm and non-suicidal self-injury

• Some examples of instruments to measure these behaviours

• More core concepts in suicide research and how to measure

• Repeated measures, time windows

• Ecological momentary assessment

Posner et al, Am J Psychiatry 2011 https://doi.org/10.1176/appi.ajp.2011.10111704

Self-Harm (SH)

Non-Suicidal Self-Injury NSSI

Suicidal Self-Harm

SuicideSuicideAttempt

• Evaluates the 6 criteria (A-F) for DSM-V NSSI disorder

• First: Administer a 17-item self-report questionnaire – Deliberate Self-Harm Inventory (DSHI) for Criterion A

• Second: Conduct structured interview for Criterion A and the rest of the criteria

Gratz, K.L., Dixon-Gordon, K.L., Chapman, A.L., & Tull, M.T. (2014)

Klonsky, E.D. & Olino, T.M. (2008).

www.selvmord.no

Outline• Clarity of concepts and definitions – why is this important?

• Defining suicide, attempted suicide, self-harm and non-suicidal self-injury

• Some examples of instruments to measure these behaviours

• More core concepts in suicide research and how to measure

• Repeated measures, time windows

• Ecological momentary assessment

www.selvmord.no

Outline• Clarity of concepts and definitions – why is this important?

• Defining suicide, attempted suicide, self-harm and non-suicidal self-injury

• Some examples of instruments to measure these behaviours

• More core concepts in suicide research and how to measure

• Repeated measures, time windows

• Ecological momentary assessment

Ecological Momentary Assessment (EMA)• AKA Event Sampling Methodology

• Repeated collection of data via mobile devices on subjects' current behavioursand experiences in real time, in participants' natural environments.

• Minimizes recall bias

• Maximizes ecological validity

• Allows more careful study of processes influencing behaviour in real-world contexts

• For example, if you aim to study mechanisms of change during treatment

• Or you wish to study highly volatile phenomena such as suicidal ideation or hopelessness or frequently occurring events such as NSSI

Ecological Momentary Assessment (EMA)• Example from my own lab: MinEMA (’MyEMA’)

• App is password protected (!)

• The app will prompt participants to respond to a set of questions six times daily (between 10 AM and 10 PM) for seven consecutive days, yielding data from a maximum of 42 measurement points

• Each data collection takes 2-3 minutes

• Data are delivered directly and fully encrypted to and storedin the project database in the dedicated project area withinour research server

Allows us to study (examples)....

• ... sequential orders – what comes first?

• ...temporal patterns in more detail – in what situations or times of day?

• ...individual / group patterns

• ...changes in patterns over time and between intervention groups

• and many more

suicideresearchsummit.org

Defining Core Concepts and Measurements in Suicidology

Jill Harkavy-Friedman, PhD

PlanBasic measurement considerationsSuicide specific considerations

What is the variable of interest?Based on level of interest and literature

• ideation, plan, intent, behavior, death• knowledge, attitude, skill, behavioral change

Level of analysis• person, family, institution, population• candidate genes/genome screen

Absolute value or change score• reduction, response, recovery

Multiple measures vs. single measure• data reduction, redundancy

Administration ConsiderationsFormat

• Face-to-face interview, self-report, behavioral observation, telephone, computer, biological

Source of information• Self, parent, other informant, observer,

records, epidemiological informationInstrument for repeated measures

• Same form, alternate forms

Who should measure?Self-report vs. other reportClinical vs. lay ratersOpen vs. blind measurementTechnician vs. computer/lab equipmentInvestigator

How to decide on a measureReliabilityValiditySensitivitySpecificityVariabilityCeiling and Floor effects

Reliability=ReproducabilityInter-rater

Kappa

Intra-class correlation

Test-Retest: Over time

Correlate time 1 and time 2

Parallel Forms: Across Measurements

Correlate forms

Internal Consistency: Within a test

Spearman BrownCronbach’s Alpha

ValidityFace Validity: Does it look like it measures what it is supposed to ?Content Validity: Is the content representative?Criterion Validity: Predictive, ConcurrentConstruct Validity: Accrual of meaning through convergent and discriminant validity

Reliability is the upper limit of validity

Can you find an effect?Sensitivity and Specificity

Variability

Ceiling and Floor effects

Determine Goal of AssessmentSuicidal ideation and behaviorRiskTreatment effectPopulation risk

No matter the goal, suicide is complex, and you will likely have to measure multiple variables

Variables for measurement• Suicidal Behavior: Ideation, attempts, completion details

• Clinical Measures: diagnosis, clinical characteristics, mood

• Psychological measures: depression, hopelessness, impulsiveness, emotion regulation

• Social History: trauma, stress, social functioning, school experience

• Cognitive functioning: decision-making, implicit bias, • Psychophysiological measures: HRV, GSR, EEG• Biological measures: neurotransmitters, hormones,

metabolomics, inflammation, gut biome• Environment: access to means,

support, housing, food security

What needs to be measured?DemographicsSuicidal ideation and behaviorOutcomeConfoundersMediators and ModeratorsContext

Knowledge # crisis callsAttitudes Associated symptomsSuicidal Ideation Impact of suicideSuicide Attempts HospitalizationCompleted Suicide School completionLethality of attemptSuicide Intent

# referralsSocial Skills

Current Measures of Outcome

For Intervention studies outcomes must:Measure the target of interventionBe standardizedBe “not average” at baselineBe expected to change within the time frameBe Sensitive to changeBe present in all groupsHave a measurable effect sizeHave demonstrated reliability and validityBe feasible

Measurements, observations, descriptions can only be considered scientific when they are independently confirmed by other people.Jose Padilha

Thank You!

suicideresearchsummit.org

@afspnational

suicideresearchsummit.org

Clinical Trial Methods: Specific Considerations

for Suicide Research

Gregory K. Brown, PhDBarbara Stanley, PhD

Common design questions to consider when conducting clinical trial research with at risk samples

• What is the research question (study hypotheses)?• What is the study intervention and how does it lower risk (mechanism)?• Is the intervention safe?• How will you know if the intervention was provided as indicated?• Who is eligible to receive the intervention?• What is the outcome domain?• How will you measure the outcome?• What is the control intervention?• How many participants will you need?• Is the study feasible?

Choose an Appropriate Suicide Outcome Domain

Suicide

Suicidal Behavior

Suicidal Ideation

Use an Established Nomenclature of Suicidal Behavior

Crosby, A. E., Ortega, L., & Melanson, C. (2011). Self-directed Violence Surveillance: Uniform Definitions and Recommended Data Elements (Version 1.0). Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control.

Self-directed Violence Surveillance: Uniform Definitions

Columbia Suicide Severity Rating Scale (C-SSRS) DefinitionsPosner, K., Brown, G. K., Stanley, B., Brent, D. A., Yershova, K. V., Oquendo, M. A., … Mann, J. J. (2011). The Columbia–Suicide Severity Rating Scale: Initial Validity and Internal Consistency Findings From Three Multisite Studies With Adolescents and Adults. American Journal of Psychiatry, 168(12), 1266–1277.

Avoid terms that are infrequently used or poorly defined: “suicide gesture” or “suicidality”

Suicide as Outcome

• Pros• High ecological validity• State and national datasets are available: National Death Index, National

Violent Death Reporting System

• Cons• Suicide is a rare event even among high risk populations and requires very

large samples• Ascertaining death by suicide can take a long time• Discerning cause of death can be challenging (suicide vs accidental overdose)

Suicidal Behavior as Outcome• Pros

• May serve as a valid proxy measure of death by suicide• May be assessed by self-report, clinician interview, informant (such as a

family member) or by using medical record data such as using ICD-10 codes• Occurs more frequently than suicides but are still rare events unless high risk

samples are used• Cons

• May be especially rare events among older populations who often kill themselves on the first attempt

• Suicidal attempts can be difficult to identify (questionable lethality/potential lethality or questionable intent to die)

• Importance of using blind assessors to prevent biased assessments• Often difficult to maintain the blind

Suicidal Behavior as Outcome• Determine the types of suicidal behavior to assess:

• Suicide attempts• Interrupted attempts?• Aborted attempts?• Preparatory behavior toward imminent suicide behavior?• Is an ED visit for a suicide-related concern a positive or negative outcome?

• Use validated measures of suicidal behavior that correspond to the nomenclature

• See PhenX Toolkit• Consider value of Common Data Elements so that data can be harmonized across

studies• Establish interrater reliability; consider using blinded adjudication boards

for difficult to classify behaviors

Suicidal Ideation as Outcome

• Pros• Often more frequent than suicidal behavior• May be assessed by clinical interview or self-report• Validated measures of suicidal ideation are available

• See Phenx Toolkit or recent reviews• Consider measures for the appropriate age group

• Severity of suicidal ideation can be classified: (wish to die, active suicidal thoughts, active suicidal thoughts with general method, suicidal intent, suicidal intent with plan)

Suicidal Ideation as Outcome

• Cons• Suicidal ideation can be highly variable over time

• Fleeting, short or long duration, or can be chronic• Subject to recall bias if assessed retrospectively

• Consider “real time monitoring” such as Ecological Momentary Assessments• Secondary gains (or loses) can influence self-report

• Desires hospitalization for reasons other than suicide risk; fears hospitalization or loss

Frequency Distribution of the Scale for Suicide Ideation (Current) During Follow-up

Data from Brown et al., JAMA, 2005

Frequency Distribution of the Scale for Suicide Ideation (Worst) During Follow-up

Data from Brown et al., JAMA, 2005

Determining Inclusion Criteria

• Measurement of Suicidal Ideation• Consider enrolling patients with history of ideation

• What is the timeframe for the ideation: day/hour of assessment, past week, past month, or lifetime?

• Use clearly defined and reliable threshold for severity of ideation• Use measures with evidence-based cut-off scores or validated types of ideation• Avoid vague or unreliable nomenclature: “significant suicidal ideation”

Sample Size Considerations – Suicidal Behavior

• Need to estimate the rates of behavior during follow-up in the intervention condition and the control condition

• Larger sample sizes are needed when measuring suicide behavior to determine if the intervention prevented the behavior thanwhen using measures to assesses changes in severity (such as continuous depression scales)

Consider Recruitment Feasibility when Establishing Thresholds on Suicide Measures

Increased Sample

Availability

Lower Power to Detect Effects

During Follow-up

Lower threshold Such as longer

timeframe

Decreased rates of suicide

behavior

Managing Participants in Suicide Intervention Trials

• What to monitor• How to monitor• What should be done if risk is detected?• Participants in the control condition---what is an adequate control?

Managing Risk Occurs throughout the Trial

• Points of managing risk• Initial contact—screening phone contact • Between screening and in-person visit/consent• Prior to randomization• During trial• Disposition

• Procedures are similar throughout• Control conditions can vary but monitoring of risk should be the same

across conditions

Tension between Safety Procedures and “Best” Research Methods• Prior to discussing Safety procedures, important to discuss their

impact on methods throughout the trial • This impact has to be considered carefully• Balance between safety and methods that will answer the research

question being asked

Case Example: Trial Comparing Paroxetine with Placebo in Suicide Attempters

• Research question—is paroxetine effective in preventing the recurrence of suicide attempts

• Outcome---Suicide attempts• Trial length---12 months• Safety procedures: Remove if suicide attempt occurs during trial

or if suicide ideation becomes “too significant” • Minimizes risks• May lead to inaccurate conclusions from biased withdrawal• May not be clinically necessary

“The time from baseline to first recurrence of a suicide attempt was considered to be the primary endpoint.”

“Paroxetine appeared to be effective in the prevention of recurrent suicide attempts. This effect was observed (only) in minor repeaters.” Verkes et al. 1998

What should be done if suicide attempts occur during a trial?•Removal and referral•Maintain in trial with standard monitoring procedures

•Maintain in trial with increased monitoring in place

What should be monitored with suicidal participants?• Suicide risk---but how is it determined?

• Increasing suicidal ideation• Level of ideation that we typically identify as problematic—ideation with

intent; ideation with intent and plan• Suicidal behavior• Increasing symptoms associated with suicidal behavior e.g.

depression, hopelessness• Need to define and set criteria at the outset of the trial• Need to set procedures for monitoring at the outset• Need to define what will be done if criteria are met

Defining and intervening on suicide risk

• This is may seem simple but it is not • Has an impact on:

• Participant safety if too minimal• Participant willingness to disclose if too strict• Study outcomes if occurs too frequently or at too low a bar

• Why do we care if safety is at stake if study outcomes are adversely affected?• Participants may endure a trial for no reason; waste of time, money and possible risk

exposure

• Obtaining risk by: 1. asking participants directly; 2. monitoring how they are responding via EMA; 3. losing contact (participant stops attending appointments, stops answering calls)

Risk determination: How and by Whom

• Obtaining risk by: • 1. asking participants directly• 2. monitoring how they are responding via EMA• 3. losing contact (participant stops attending appointments, stops answering

calls)

• Who assesses risk

EMA as Tool to Measure SI: Comparison of SSI and EMA SI• Worst-point EMA ideation was positively related to the retrospective

post-EMA SSI (r=.729, p < .001)

• EMA ideation severity was also positively related to the retrospective post-EMA SSI; participants with one point higher on the post-EMA SSI had on average .85 higher scores on each EMA ideation item (SE=0.10)

• However, 58% of participants reporting ideation with EMA denied past week ideation when assessed retrospectively over the same timeframe on the SSI

Demographic and clinical characteristics by whether or not post EMA SSI = 0

Individual EMA suicidal ideation trajectories for participants with post-EMA SSI scores of zero

Comparing EMA SI and SSI=0

Mixed effects model of EMA suicidal ideation item endorsement on having non-zero post-EMA SSI

Note. + items were reverse-coded.

EMA Monitoring and Intervening

• Sometimes we do not know enough about when to intervene• Intervening can have a significant impact on future responding making the

assessment meaningless• Suggested approach---to monitor EMA remotely

• How often will EMA be monitored? Daily? 24/7? • Suicidal crises often last only minutes to a few hours in escalation from ideation to attempt• Identify point at which intervention will occur• Determine what will the intervention be

• Alternative approach---no EMA monitoring • Inform participants that EMA will not be monitored; that it is not a communication method• Provide emergency contact information as you would if assessments were done in the usual

way—clinician interviews, weekly self ratings

Case Example: Real time monitoring studies of suicidality: When to intervene

In the past 15 minutes, how strongly have you felt or experienced the following:

1. A wish to live 0 1 2 3 4

2. A wish to die 0 1 2 3 4

3. A wish to escape 0 1 2 3 4

4. Thoughts about dying 0 1 2 3 4

5. Thoughts about suicide 0 1 2 3 4

6. Urge to commit suicide 0 1 2 3 4

7. Thoughts about hurting self 0 1 2 3 4

8. An urge to hurt yourself 0 1 2 3 4

9. Like there were reasons for living 0 1 2 3 4

Individual with Highly Variable Suicidal Ideation: When to Intervene?

Individual with Elevated, Stable Suicidal Ideation

Participant Safety Procedures• Phone contacts---at beginning obtain phone number to recontact and

physical location• Develop a safety plan---clinical tool• Have full discussion with participants about emergency procedures

with study staff should they become suicidal—research tool• Provide a written document with emergency procedures and study-specific

contact information • Set the stage where investigators encourage rather than discourage contact if

participants are struggling• Suicide risk should be assessed clinically on a routine basis in addition

to study assessments

Staff Safety Procedures• All staff should have specific risk assessment training• Specific safety procedures should be clearly laid out for all study staff• For phone interactions, staff should have a way to connect with

senior staff or emergency rescue without ending the call with the participant

• A senior investigator should always be available to assessors and research assistants for consultation; set the stage---better to consult than try to handle matters alone; let staff know to say that they have an emergency

Emergency Procedures• Obtain emergency contacts at time of enrollment and permission to

use them• Identify conditions to participants when you will use emergency

contacts• Describe limits to confidentiality—if imminent suicide risk,

confidentiality cannot be maintained• This discussion takes place during consent process but it is good to reinforce

this periodically so participants are not surprised • Describe emergency rescue procedures and how collaboration and

cooperation can mitigate their use• Transparency is crucial

Postvention

• Establish procedures in advance should a suicide or highly lethal attempt occur during the trial

• Identify to whom events are reported

• Provide support for staff• Determine how contact with family will be handled

Final Points to Consider

• Ensure adequate staff time• Ensure adequate funding• Ensure support for all staff including senior investigators• Use consultation with peers extensively• Keep in mind that the work is hard but the goal is extremely

rewarding• Safety planning feedback from users

The Pathophysiology of

Suicidal Behavior

J. John Mann, MD

IASR/AFSP Workshop 2020

Disclosures:1. This talk is based on research funded by NIMH & BBRF.

2. Recipient of royalties from Research Foundation for

Mental Hygiene for commercial use of the C-SSRS.

A Brain-Centric Model of Suicidal Behavior: Mann and Rizk, AJP 2020.

J Mann 2020

Stress Diathesis Model of Suicidal Behavior

J Mann 2020

External

Stressful Life

Events

Perception

of Stress

(depression

& social

cues)

Response

to Stress

Internal

Stress of

Major

Depression

INSERT PHOTO

Milak et al, J Affective Disorders, 2010

Subjective depression, and

clinician-rated depression are

associated with different brain

regions.

Subjective Depression Associated with Anterior Cingulate Cortex Hyperfunction and

dlPFC Hypofunction

Brain Blood Flow Predicts Suicide in Major Depression

J Mann 2020

Dorsolateral PFC and

insula hypofunction are

seen in future suicides.Willeumier et al Trans

Psychiatry (2011)

Responses to Emotional Faces in Euthymic Suicide

Attempters versus Nonattempters Show Social Distortion

Jollant et al, AJP 2008, 165

Stress Diathesis Model of Suicidal Behavior

J Mann 2020

External

Stressful Life

Events

Perception

of Stress

(depression

& social

cues)

Response

to Stress

Internal

Stress of

Major

Depression

Delayed Discounting

* Value of rewards are discounted in proportion to delay.

* Value of uncertain rewards are even more discounted.

* Degree of discounting is a trait.

* Delayed discounting is an unconscious mechanism.

J Mann 2020

Clinical implications for decision to die by suicide or not?

• Suicide offers immediate certain relief from pain associated

with life.

• Treatment offers uncertain future benefit.

• Treatment is a harder sell to a patient prone to delayed

discounting and because of uncertainty of response.

J Mann 2020

A Revised Model of Decision Making and Suicidal Behavior

J Mann.2020

Emotional

painValue of relief by suicide or relief by

antidepressant treatment

Treatment

Low lethality impulsive suicidal behavior High lethality planned suicidal behavior

Need for rapid relief even if survives attempt or value of

certainty of death

Brain Blood Flow Predicts Suicide in Major Depression

J Mann 2020

Dorsolateral PFC and

insula hypofunction is

associated with

severity of subjective

depression and more

pronounced in future

suicides.Willeumier et al Trans

Psychiatry (2011)

Dorsolateral PFC Regulates Risk-taking Behavior

* Healthy men, increase risk-taking choices on a gambling task

when transcranial magnetic stimulation inhibits dorsolateral

PFC presumably because top down effect on orbital PFC is

compromised (Knoch et al 2006).

* Imaging of MDD at risk for suicide shows hypoactive dlPFC.

* Dorsolateral PFC impaired > orbital PFC > risky decisions

and suicidal behavior

J Mann 2020

Impaired Learning During Iowa Gambling Task by Suicide Attempters:

failure to improve problem solving

Jollant et al. AJP, 2005

Neurobiology of Suicide: seven pathways

1. High 5-HT1A autoreceptors > low serotonin release> low activity>loss of trophic effect2. Low CSF MHPG = low noradrenergic activity3. Low GABA = low GABAergic activity4. High glutamate>neurotoxicity5. High HPA axis activity>neurotoxicity6. Inflammation>neurotoxicity7. Low omega 3/6 PUFA ratio, stress>neuroinflammationand altered brain activity/neurotoxicity

J Mann.2020

Stress and Inflammation

• Inflammation is how the body defends against

infection and cancer.

• Inflammation is how the body repairs after trauma.

• Inflammation is a response to stress.

J Mann.2020

Inflammation in the Brain

• Inflammation outside the brain affects the brain and produces “sickness” behavior or state.

• Inflammation in body can cross the BBB and affect the brain by producing inflammation in the brain.

• Infections can cross the BBB and produce inflammation in the brain.

• COVID-19 has not been shown convincingly to get into the brain but does affect brain blood vessels and cause strokes.

J Mann.2020

Inflammatory Response is Triggered by Emotional Stress

J Mann.2020

A PET Scan of Inflammation in Brain: TSPO binding

J Mann.2020

ER176 VT (not adjusted for genotype)

BD

I

anterior cingulate orbital PFC medial PFC

y = 6.991x - 4.672R² = 0.692

y = 7.695x - 6.943R² = 0.691

y = 7.569x - 4.888R² = 0.692

0

5

10

15

20

25

30

35

40

0 2 4 6 8

J Mann.2020

ER176 VT (not adjusted for

genotype)

SU

ICID

AL

ID

EA

TIO

N

anterior cingulate orbital PFC medial PFC

y = 1.497x - 2.378R² = 0.179

y = 1.875x - 3.787R² = 0.231

y = 1.709x - 2.763R² = 0.199

0

2

4

6

8

10

12

14

0 2 4 6 8

J Mann.2020

Stress Diathesis Model of Suicidal Behavior

J Mann.2020

External

Stressful Life

Events

Perception

of Stress

(depression

& social

cues)

Response

to Stress

Internal Stress

of Major

Depression

Dranovsky and Hen, 2006:

Stress in mice > fewer cells and smaller cells in hippocampus

Antidepressants > more and bigger cells

J Mann.2020

More Time in a Major Depression Produces Smaller Hippocampus

Sheline et al PNAS, 1996

Antidepressants Appear to Correct

Dentate Gyrus Volume Deficit in Depression

n=18 n=18 n=8 n=5 n=4

p<.001

Boldrini et al, BP 2012

Fewer Mature Neuronal Granule Cells in Dentate Gyrus in

Untreated MDD Suicides.

Boldrini et al, BP 2012.

SSRI-Treated

MDD Are Same as

Controls

Process Length/Synapses In MDD Suicides

Boldrini et al, unpublished

50 μm

A

Hilus

ML

SGZ

CA3

GCL

500 μm

B

C

GCL

GCL

Hilus

ML

SGZ

Shorter dendrite length in anterior DG in suicide-MDD

200

400

600

Den

dri

te len

gth

(u

m)

*p=.006

Control

0Suicide

MDD

100 μm

Process retraction in MDD suicides indicates synapse loss

Shorter Serotonin Neuron Process Length in PFC of Suicide Decedents

Austin et al. Neuroscience 2002

Process length is shorter in some

layers of Brodmann Area 46 in dlPFC.

Brain BDNF Lower in Depression and Suicide If History Of

Childhood Adversity

J Mann 2020

C A 3 #

Anterior HPC

p=.030

5-H

T1A

mR

NA

de

ns

ity

(n

Ci/m

g tis

su

e)

C –No ELA

C –ELA

MDD –No Suicide

MDDSuicide –No ELA

MDDSuicide –

ELA

0

0.1

0.2

0.4

0.3

Figure 5. 5-HT1A receptor mRNA in situ hybridization in

hippocampus from subjects with and without childhood adversity(ELA). 5-HT1A receptor mRNA is more in anterior DG in suicide vs

non-suicideMDD.

*

Nonsuicide, No Adversity Suicide, No Adversity Nonsuicide, Adversity Suicide, Adversity

Figure 6. FKBP5 Levels in Prefrontal Cortex.

Subjects with a history of early life adversity havehigher FKBP5 levels in both dorsolateral and

cingulate prefrontal cortex.

BA24BA9

Figure 7. HDAC levels in prefrontal cortex. In dorsolateral prefrontal

cortex (BA 9) Suicides with early life adversity had reduced HDAC6levels compared to suicides with no history of early life adversity.

In anterior cingulate cortex (BA 24) suicides had lower HDAC6levels than nonsuicides, independent of adversity status.

0.5

0

*

BD

NF

:HK

P r

ati

o

2.0

1.0

1.5

1.5

1.0

0.5

0

*

*

BA 24BA 9

0.8

0.6

0.4

0.2

0.0

* *

HD

AC

6:H

KP

ra

tio

Figure 8. BDNF Protein. Western blots

were analyzed by autoradio-graphy.Both suicide groups as well as controls

exposed to adversity had lower ratiosthan controls.

BA 24

FK

BP

5:H

KP

rati

o

p<.05

HPA Axis Over-activity and Neuroinflammation in Suicide

J Mann 2020

CA3#

*

Nonsuicide, No Adversity Suicide, No Adversity Nonsuicide, Adversity Suicide, Adversity

Figure 6. FKBP5 Levels in

Prefrontal Cortex. Subjects witha history of early life adversity

have higher FKBP5 levels inboth dorsolateral and cingulate

prefrontal cortex.

BA24BA9

Figure 7. HDAC levels in prefrontal

cortex. In dorsolateral prefrontal cortex(BA 9) Suicides with early life adversity

had reduced HDAC6 levels compared tosuicides with no history of early life

adversity. In anterior cingulate cortex (BA24) suicides had lower HDAC6 levels than

nonsuicides, independent of adversitystatus.

0.5

0

*

BD

NF

:HK

P r

ati

o

2.0

1.0

1.5

1.5

1.0

0.5

0

*

*

BA 24BA 9

0.8

0.6

0.4

0.2

0.0

* *

HD

AC

6:H

KP

ra

tio

Figure 8. BDNF Protein.

Western blots wereanalyzed by autoradio-

graphy. Both suicidegroups as well as

controls exposed toadversity had lower

ratios than controls.

BA 24

FK

BP

5:H

KP

rati

o

p<.05

1.5

Cyto

kin

es

LeptinSCFMIG

MIP1AMCP3

PAI1FASL

ENA78IL1B

IL2IL4IL5

IP10TGFA

IL6IL7IL8

IL10TGFBIFNB

TNFBIL12P40IL12P70

IL13IL17

RANTESIFNG

GMCSFTNFAGCSF

MIIP1BIFNA

LIFMCP1

EOTAXINGFGBFEGFTRAILGROA

IL1AIL1RA

IL15ICAM1

HGFCD40L

RESISTINVCAM1

MCSF

PDGFBBNGF

IL17F

1.00.50.0-log10(p)

Neuroinflammation in Suicide MDD

HPA Axis Overactivity

Trophic Deficits and Toxic Effects in MDD Suicides

• Lack serotonin/BDNF trophic effects.

• Excessive HPA allostatic load.

• Neuroinflammation.

• All favor process and cell loss.

J Mann.2020

Summary

• Brain function is abnormal in high suicide risk patients and decedents in brain areas related to emotion regulation, social perceptions, decision-making and learning.

• Stress raises HPA activity, increases inflammation and lowers BDNF.

• Inflammation reduces serotonin function.• All reduce processes and cell survival and increase risk

of suicide.

J Mann.2020

Genetics and Epigenetics in Suicide Research

Gil Zalsman MD, MHA

President of the IASR

Director of Geha MHC

Chair of Psychiatry, Tel Aviv University, Israel

and Molecular Imaging and Neuropathology Division,

Columbia University, USA

Basic Principals

Chapters: 10-14

Specific for suicide

Suicide runs in families

Suicide runs in families(A Roy et al 1990, DA Brent et al., 1996)

Ernest Hemingway

Suicide runs in families

A. Families studiesWhat we do using this method?

• Familial aggregation of suicidal behavior

• Assessing relatives of attempters/died by suicide

• Population registry in Denmark and Sweden (Asberg 2003, Qin 2002)

Strengths and weaknesses of this method

• Most are retrospective

• Environment confounders? No just due to grief (sui>homicide)

• Prospective- lots of years and resources, government will…

Suicide runs in families

B. Adoption studiesWhat we do using this method?

• Using adoption registry

• Matching adopted subjects who died by sui to those who didn’t looking

at their biological vs non biological parents (Schulsinger 1979)

• Controls for environmental confounders

Strengths and weaknesses of this method

• Needs open registry (Denmark)

• Most are retrospective

• Prospective- lots of years and resources, government will…

Suicide runs in families

C. Twins studiesWhat we do using this method?

• Using twins registry

• MZ vs DZ

• Evaluate magnitude of gene vs environment effects

• Twins registry in Denmark

Strengths and weaknesses of this method

• Needs registry

• Shared and non-shared environment

• Most are retrospective

• Prospective- lots of years and resources, government will…

Suicide runs in families

C. Twins studies

DZ 0.7%

MZ 13%(Roy A. 1990; Ott J. et al, 2001)

Caspi and Moffitt, Nature Reviews Neuroscience, July 2006, with permission

Approaches in research of the medical genetics

Association studies in suicidology

Association Studies in SuicidologyWhat we do using this method?

• Assessing specific candidate genotype frequency in affected vs non affected

subjects. Can use intermediate phenotypes (endophenotype)

• Assuming direct main effect by a single allele/SNP/polymorphism

• Chi square statistics

Strengths and weaknesses of this method

• Looking under the light

• Association doesn’t mean effect or causality

• If there is a direct main effect of a single marker it’s a Nobel price…

• Simple PCR technique (learn how to)

• Needs large numbers (n)

• Needs good clinical phenotype (Questionnaires) OR ENDOPHENOTYPE

• Environmental effects are not accounted for

e.g.5HTTLPR

Genotype Non suicidal

(expected)

Suicidal

(finding)

LL 80% 20%

SS 15% 75%

SL 5% 5%

Direct main effect approach

TPH1

TPH2

SERT-5HTTLPR

COMT val/met

MAO A

5HT’s

DRD4

NET

BDNF

Wolfram (WFS1)

Etc……

→ Equivocal results

Haplotype Relative Risk (HRR)

TDT

Parents are controls for their suicidal kid

*Avoid Ethnic Stratification

Transmitted alleles Non-transmitted allele

AA BA

AA BA

HRR association approach

Genetics of Suicide in Adolescents

Zalsman G

In: Dwivedi Y, editor. The Neurobiological Basis of Suicide. 2012. Chapter 14.

GWAS in suicidology

Genome Wide Association Studies

What we do using this method?

• Multiple association studies in one shot

• DNA microarrays

• RNA expression arrays

• Looking for linkage between specific SNPs and suicide phenotypes

Strengths and weaknesses of this method

• Needs large numbers (n)

• $$$$ (not anymore)

• University setting-genome center

• Environmental confounders

• Multiple testing: many SNPs are very significant….Use post hoc tests

(e.g. Hochberg-Binayminy) or look for candidate SNPs

Genome Wide Association Studies

Just came out: Ducherty et al.,

AJP October 2020

the first comprehensive genomic analysis of suicide

death using previously unpublished genotype data

from a large population-ascertained cohort.

Genome-wide association analysis identified

two genome-wide significant loci (involving six

SNPs: rs34399104, rs35518298, rs34053895,

rs66828456, rs35502061, and rs35256367).

Gene-based analyses implicated 22 genes on

chromosomes 13, 15, 16, 17, and 19 (q<0.05).

Polygenic scores for several other psychiatric

disorders and psychological traits were also

predictive, particularly scores for behavioral

disinhibition and major depressive disorder.

Caspi and Moffitt, Nature Reviews Neuroscience, July 2006, with permission

Approaches in research of the medical genetics

GxE approach in suicidology

Binder 2016

5HTTLPR Gene X Environment Interaction in suicidal behavior

Caspi et al. Science, 2003

GxExT approach

Suggested Model: GxExT interaction

WKY Rat

Animal model for depression,

despair and anhedonia

GXEXT

Zalsman et al., Eur Neoropsychopharmacology 2015

Epigenetics

EpigeneticsChanges in DNA that change gene expression. These

changes can be permanent (cell type) or temporary

(developmental window , environmental ques)

Types:

1. Methylation

2. Histones modification

3. Non coding RNAs=MiRNA

Issler and Chen, Nature Review Neuroscience 2015

Epigenetics in Psychiatry

Epigenetics in suicidology

Labonte 2013

Caspi and Moffitt, Nature Reviews Neuroscience, July 2006, with permission

Approaches in research of the medical genetics

How environment and epigenetics interact?

epigenomic marks can be altered through calcium-dependent signaling cascades in direct response to neuronal activity.

Nagy C et al., Genes Brain Behav. 2018;17(3):e12446.

Thank You!zalsman@tauex.tau.ac.il

Thank You!zalsman@tauex.tau.ac.il